[J3] work on F202X at #217?

Clune, Thomas L. (GSFC-6101) thomas.l.clune at nasa.gov
Tue Jul 24 08:27:23 EDT 2018


Anton,

OK - I’ll wade in this time.  My excuse being ignorance which gives me some cover for being wrong.

There are no “rules” in the sense that you are seeking, though there certainly is experience (and baggage) among the existing committee members in terms of what can work.     (Not me - I joined after this phase of the previous cycle.)   I would very much like to have at least a rough sketch of the proposed syntax by the time we leave the next meeting.   That will give us time to exercise the syntax against the use cases and for the vendors to further analyze the implementation difficulty.   If a change of course is needed, then we have one more meeting before the next WG5.

With regard to generics,   I feel there needs to be some degree of consensus on the overall approach before we can seriously consider syntax.     Depending on that consensus, we may just be modifying (tweaking?)  an existing proposal or we may be starting from scratch.  For myself, I’m trying very hard to be a fence-sitter, as I have yet to hear Malcolm’s insights on this issue.  Others have shared his previous position at a high level, but not his rationale.       

I fear you may be using this issue to decide on whether to attend the next meeting in Vegas.   If so, there will be some degree of risk either way.   

Cheers,

- Tom




> On Jul 24, 2018, at 6:49 AM, Anton Shterenlikht via J3 <j3 at mailman.j3-fortran.org> wrote:
> 
> Bill
> 
> On Fri, Jul 20, 2018 at 10:46:13PM +0000, Bill Long via J3 wrote:
>> Our primary task is to provide more specific
>> proposals for WG5 to consider for the F202X
>> content at the Tokyo meeting in 2019.
> 
> How much "more specific"?
> 
> Take generics/templates as an example.
> A number of papers, including
> use cases, have been submitted
> for 215, e.g. Tom's, Van's, yours.
> A lot of discussion of these
> already happened at 215, so what
> exactly is the procedure for developing
> this proposed new feature for WG5?
> Plenary and subgroup discussions
> of the relative merits of different
> proposals and how they address the use
> cases?
> 
> Is the aim to have a single agreed
> proposal for this feature for WG5?
> Or 2-3 different proposals for WG5 to
> discuss and vote on?
> 
> Is the aim to have a sufficiently
> detailed proposal on which the users
> can comment in time for 218? 
> 
> Will there be any proposed syntax in 217,
> or just a description in English of how
> the new facility will work?
> 
> Sorry for potentially stupid questions.
> Perhaps this is all clear to those who've
> been there at the beginning of 2018, 2008, 2003, etc.
> 
> Thanks
> 
> Anton



More information about the J3 mailing list